APHASIA IN ACUTE STROKE/flrusf et al. 167

31. Gotoh F, Maramatsu F, Fukuuchi Y, et al- Dual control of cerebral cir-culation. Separate sites of action m vascular tree in autoregulation andchemical control, (abstr) Stroke 4:327, 1973

32. Thierry AM, Stinus L, Blanc G, et al' Some evidence for the existence ofdopaminergic neurons in the rat cortex. Brain Res 50:230-234, 1973

33. von Essen C: Effects of dopamine on the cerebral blood flow in the dog.Acta Neurol Scand 50:39-52, 1974

34. Garbarg M, Barbin G, Feger J, et al: Histaminergic pathway in rat brainevidenced by lesions of the medial forebrain bundle. Science 186:833-835,1974

35. Sokoloff L: The action of drugs on the cerebral circulation. PharmacolRev 11:1-85, 1959

36. Shenkin HA' Effects of various drugs upon cerebral circulation andmetabolism of man. J Appl Physiol 3:465-471, 1951

37. Chasin M, Rivkin 1, Mamrak F, et al: Alpha- and beta-adrenergic recep-tors as mediators of accumulation of cyclic adenosine 3',5'-monophosphate in specific areas of guinea pig brain. J Biol Chem246:3037-3041, 1971

38. Peerless SJ, Yasargil MG, Kendall MJ The adrenergic and chohnergicinnervation of the cerebral blood vessels. Excerpta Med 272:199-202,1971

39. Deshmukh VD, Meyer JS, Aoyagi M, et al- Evidence for chohnergicneurogenic influence on cerebral blood flow, (abstr) Stroke 6:226, 1975

40. Fuxe K, Hokfelt T, Ungerstedt U: Morphological and functional aspectsof central monoamine neuron. Int Rev Neurobiol 13:93-126, 1970

41. Welch KMA, Hashi K, Meyer JS: Cerebrovascular response to in-tracarotid injection of serotonin before and after middle cerebral arteryocclusion. J Neurol Neurosurg Psychiat 36:724-735, 1973

42. Ekstrom-Jodal B, von Essen C, Haggendal E, et al: Effects of 5-hydro-xytryptamine on the cerebral blood flow in the dog. Acta Neurol Scand50:27-38, 1974

43. Welch KMA, Meyer JS, Chee ANC: Evidence for disordered cyclic

AMP metabolism in patients with cerebral infarction. Europ Neurol13:144-154, 1975

44. Edvinsson L, Nielsen KC, Owman Ch: Sympathetic nervous influence onthe blood circulation and carbonic anhydrase activity in the choroidplexus, (abstr) Stroke 4:368, 1973

45. Kotler MN, Berman L, Rubenstein AH: Hypoglycemia precipitated bypropranolol. Lancet 2:1389-1390, 1966

46 Goodman L, Gilman A (eds): The Pharmacological Basis ofTherapeutics. New York, MacMillan Co. Chap 26, p 569, 1970

47. Burns TW, Langley PE: Differential effects of alpha- and beta-adrenergicblocking agents on basal and stimulated lipolysis of human and ratisolated adipose tissue cells. Pharmacol Res Comm 3:271-277, 1971

48. Burns TW, Mohs JM, Langley PE, et al: Regulation of human lipolysis.In vivo observations on the role of adrenergic receptors. J Clin Invest53:338-341, 1974

49. Meyer JS, Ryu T, Toyoda M, et al: Evidence for a Pasteur effectregulating cerebral oxygen and carbohydrate metabolism in man.Neurology 19:954-962, 1969

50. Pendleton RG, Newman DJ, Shelman SS, et al. Effect of propranololupon the hemoglobin-oxygen dissociation curve. J Pharmacol Exp Ther180:647-656, 1972

51. Leonard BE The effect of some alpha-adrenergic blocking and otherdrugs on brain lactate levels following electroshock. Neuropharmacol10:517-520, 1971

52 Mahler DJ, Humoller FL Effects of serotonin on brain metabolism IntJ Neuropsychiat 3:229-233, 1967

53. Williamson JR- Metabolic effects of epinephrine in the perfused ratheart. II. Control steps of glucose and glycogen metabolism. Mol Phar-macol 2:206-220, 1966

54. Itaya K, Ui M: The inhibitory action of serotonin on free fatty acidutilization by rat-mesentenc adipose tissue. Biochim Biophys Acta84:604-606, 1965

Aphasia in Acute Stroke

JOHN C. M. BRUST, M.D.,« STEPHEN Q. SHAFER, M.D.,t

RALPH W. RICHTER, M.D.,$ AND BERTEL BRUUN, M.D§

S U M M A R Y Previous surveys of stroke populations have offeredonly cursory information on language disturbance, and, conversely,few surveys of aphasic populations have dealt exclusively with strokeor with acute phenomena. This paper describes aphasia in 850 acutestroke patients consecutively registered by the Harlem RegionalStroke Program, of whom 177 (21%) were aphasic; of these, ninewere of Broca's type, 24 were of Wernicke's type, 14 were anomic,ten were conduction, seven were of "isolation" type, and 107 were"mixed." An unexpected finding was a significant over-

representation of men among the nonfluent aphasics.During the following four to 12 weeks, 12% of fluent aphasics died,

and 12% remained moderately or severely impaired; among survivors,aphasia improved in 74%, and in 44% it cleared completely. Duringthe same period, 32% of nonfluent aphasics died, and 34% remainedmoderately or severely impaired; among survivors, aphasia improvedin 52%, and in only 13% did it clear completely. In both fluent andnonfluent groups, hemiparesis and/or visual field cut were associatedwith poor prognosis.

Introduction

APHASIA is a common accompaniment to stroke, and isoften the most disabling sequela. Yet surveys of strokepopulations18 have tended to offer limited information as tothe type and degree of aphasia present. Conversely, aphasia

•Director, Department of Neurology, Harlem Hospital Center, 135thStreet and Lenox Avenue, New York, New York 10037; and AssociateProfessor of Clinical Neurology, College of Physicians and Surgeons, Colum-bia University, 630 West 168th Street, New York, New York 10032.tClinical Fellow, Department of Neurology, Harlem Hospital Center.t Associate Dean and Professor of Neurology, University of Oklahoma, TulsaMedical College, Tulsa, Oklahoma. §Associate in Neurology, College ofPhysicians and Surgeons, Columbia University.

Supported in part by Health Services and Mental Health AdministrationGrant No. RM 0058 via the New York Metropolitan Regional MedicalProgram Grant.

Reprint requests to Dr. Brust, Harlem Hospital Center.

studies9'21 have seldom involved only stroke patients, orpatients acutely affected. The present study analyzes the in-cidence of aphasia types among acute stroke patients seen atHarlem Hospital, and relates the clinical findings to earlyprognosis.

Methods

From 1971 through 1973, 850 patients were admitted toHarlem Hospital with the diagnosis of acute stroke. Criteriafor inclusion in the registry of the Harlem Regional StrokeProgram have been previously described.22 Upon admissionthe patient was examined by members of the Medical HouseStaff, plus a neurology consultant. In addition, within 12 to48 hours and then at periodic intervals an examination wasperformed by a member of the Stroke Program Team andrecorded on standardized forms. The data of this report are

by guest on May 28, 2018

http://stroke.ahajournals.org/D

ownloaded from

168 STROKE VOL. 7, No. 2, MARCH-APRIL 1976

based upon review of both the Stroke Program forms andthe patients' hospital records.

Spontaneous speech was studied for fluency, articulation,the presence of paraphasia, alterations in grammar, and thedegree of comprehensibility of the patient's speech to the ex-aminer. Nonfluent speech was defined as an abnormaldecrease in the number of words used per minute, withhesitations on initiating speech and abnormalities ofrhythm; also there was usually impaired prosody, the inflec-tion and "music" of speech. Comprehensibility could belimited in severely nonfluent speech simply because of thelimited number of words used, or because of poor articula-tion. Comprehensibility could be quite adequate despitesevere nonfluency, the words sometimes consisting largely ofsubstantives delivered "telegrammaticaHy." With fluentspeech the limiting factor to comprehensibility was often thenumber of verbal or literal paraphasias, which, whendominating speech, produced incomprehensible jargon.

The patient's own comprehension of spoken speech wastested by giving commands plus asking the patient to answerobvious questions with yes or no and to indicate objectsnamed by the examiner. Ability to name was assessed forobjects, body parts, and, in many cases, colors. Ability torepeat was tested with a simple phrase such as: "Today is avery nice day," plus a syntactically loaded phrase such as:"No ifs, ands, or buts." Writing was tested mainly to dicta-tion; reading was tested aloud and for gross comprehension.

Fluency, auditory comprehension, naming, and repetitionwere judged to be normal, mildly abnormal, or severely ab-normal based on the subjective impression of the examiner.More quantitative assessment was not attempted. Aphasiawas called mild, moderate, or severe based on combinationsof these features, especially speech comprehension, plus thedegree to which the patient's speech could be understood.With Broca's aphasics severity was based upon speech out-put alone.

The aphasia classification of Geschwind23 was usedbecause of the simplicity and brevity of testing required.Broca's aphasia is defined in terms of nonfluency not ex-plained by anomia, and with comprehension preserved.

TABLE 1 Aphasia Findings in 850 Acute Stroke Patients,by Type

Finding% of specified % of all % of all

No. group aphasics pis.

Fluent aphasiaSubtype

Wernicke 24 42 13.6 2.8Conduction 10 18 5.6 1.2Anomic 10 18 5.6 1.2Isolation 7 12 3.9 0.8Uncertain 6 11 3.4 0.7Group total 57 101 32.1 6.7

Nonfluent aphasiaSubtype

Mixed 107 89 60.5 12.6Broca 9 8 5.1 1.1Anomic 4 3 2.3 0.5Group total 120 100 67.9 14.1

Total of aphasics 177 - 100 20.8No aphasia 673 - - 79.2Total pts. 850 - - 100

Other features, e.g., poor articulation, agrammatism,"recurring utterance,"24 or relative preservation ofemotional speech,25 may be present. Fluent aphasia subtypesare based on combinations of abnormal naming, com-prehension, and repetition. In Wernicke's aphasia com-prehension and repetition are both impaired to the samedegree. In anomic aphasia they are both preserved. In con-duction aphasia repetition is a good deal more abnormalthan comprehension. In "isolation of the speech area"aphasia (transcortical sensory aphasia) the reverse holds.

We are aware that this classification represents an over-simplification of a probably far more heterogeneous popula-tion, and that classifications based on broader psychologicaltesting may be closer to the truth. In a city hospital withfrequent emergency admissions a system to be useful has tobe short and straightforward.

Results

Table 1 shows that of 850 total stroke patients, 177 (21%)had aphasia acutely. In 57 (32%) speech was fluent, and in120 (68%) it was nonfluent.

In nine of the nonfluent patients, gross comprehensionwas preserved, and other features characteristic of Broca'saphasia were present. These additional features, however,varied markedly from one patient to the next. In most of theBroca aphasics naming was only mildly impaired and repeti-tion was normal. (Those with preserved repetition would beclassified by some as "transcortical motor aphasia."26)

Four patients had hesitant nonfluent speech with normalcomprehension, but here nonfluency seemed secondary toimpaired naming, other features suggesting Broca's aphasiawere absent, and the patients were therefore classified asnonfluent anomics.

One hundred seven patients had mixed aphasia with non-fluent speech plus impaired comprehension, mild in 35 andsevere in 72 (global aphasia). Of the 35 with comprehensiononly mildly impaired, naming and repetition were either nor-mal or only mildly impaired in 16; six had severely impairedrepetition and thus appeared to have combined features ofBroca's aphasia and conduction aphasia. Of the 72 mixedaphasics with comprehension severely impaired, namingwas, surprisingly, only mildly impaired in six. The rest hadeither severe impairment of both naming and repetition, orelse one or the other modality was not tested, usuallybecause the patient would not attempt the task.

Twenty-four of the fluent aphasics met the criteria forWernicke's aphasia; 13 of these were severely affected and11 only mildly affected. Ten patients had conductionaphasia; eight had severe and two mild impairment of repeti-tion, but in only one was repetition severely impaired in the

TABLE 2 Average Ages of Fluent and Nonfluent Aphasicsand of Nonaphasics

Years Range

FluentNonfluent

Broca'sAnomics

Nonaphasics

6568606265

40-10033- 9433- 7650- 7728- 95

by guest on May 28, 2018

http://stroke.ahajournals.org/D

ownloaded from

APHASIA IN ACUTE STROKE/flrt«r el al. 169

TABLE 3 Sex of Fluent and Nonfiuenl Aphasics and of Non-aphasics

TABLE 4 Hemiparesis in Fluent and Nonfiuent Aphasics

Men Women

FluentNonfiuent

Broca'sAnomicsMixed

NonaphasicsTotal

25(44%)65(54%)42

59(55%)288(43%)378(44%)

32(56%)55(46%)52

48(45%)385(57%)472(56%)

presence of normal comprehension. Ten patients had fluentanomia, mild in nine. Seven patients had "isolation of thespeech area" aphasia; in four of these there was severely im-paired comprehension, but in only one was it accompaniedby entirely normal repetition, and in none was there frankecholalia. In six of those with fluent speech, repetition wasnot assessed, and their aphasia types therefore remain uncer-tain.

As with the nonfiuent group, naming in the fluent patientsoften failed to parallel either comprehension or repetition.Moreover, the speech output within each of the fluentgroups varied widely, especially among the Wernickeaphasics, in whom it ranged from intelligible speech withpreserved patient insight to either verbal or neologisticjargon and anosognosia.

Table 2 shows no significant difference in average agebetween fluent and nonfiuent aphasics or between aphasicsand nonaphasics. Table 3, however, shows that men weresignificantly (Chi-squarecorr = 4.87, d.f. = 1, P < 0.05)overrepresented among the nonfiuent aphasics as comparedto the remainder of the 850 patients (fluent aphasics andnonaphasics). While the majority of nonaphasic patientswere women, whereas men slightly predominated amongtotal aphasics, the difference in the male-female ratiobetween these groups was not significant (Chi-squarecorr = 3.36, < 0.05 P < 0.10).

High blood pressure (defined as diastole at least 100 mmHg) was present in 53% of the aphasic patients, with nosignificant difference between fluent and nonfluent aphasicsor between aphasics and nonaphasics.

The cerebrospinal fluid was grossly bloody with xantho-chromia in 9% of the 51 fluent aphasics in whom it waschecked and in 19% of 100 nonfluent aphasics. Thedifference was not significant.

Table 4 reveals that moderate to severe hemiparesis wassignificantly (Chi-squarecorr = 41.66, P < 0.001) more fre-quent among nonfluent than fluent aphasics. A reliablehistory of handedness in the patient or his family was deter-mined too infrequently to produce meaningful data, but onepatient with nonfluent speech and mildly impaired com-prehension did describe convincing right-handedness in thepresence of acute left hemiparesis.

Homonymous hemianopia, often indeterminable becauseof lack of patient cooperation, is described in table 5. Atleast 42% of the fluent aphasics had no gross field cut, butonly two patients with severely impaired comprehensionlacked one. Seven patients with nonfluency and severely im-paired comprehension had no field cut.

As noted, aphasia severity was based on features, e.g.,speech output and speech comprehension, which did not

FluentNonfiuent

Broca'sAnomicsMixed

Total

Moderate to severe

17(30%)97(81%)7(78%)2(50%)

88(82%)114(64%)

Mild

25(44%)17(14%)2(22%)1(25%)

14(13%)42(24%)

None

15(26%)6( 5%)01(25%)5( 5%)

21(12%)

always run in parallel. Severity was graded on admission ex-cept for several patients initially stuporous or mute whobecame testable within a few days. (An interesting exceptionwas a man who remained mute up to his discharge aftermany weeks. He showed poor auditory comprehension,writing, and reading, but was cooperative and had relativelypreserved non-language mental functioning. He has been in-cluded in the mixed aphasia group.)

Table 6 records severity of aphasia on admission.Seventy-five percent of all aphasics were either moderatelyor severely impaired, and this degree of impairment wassignificantly (Chi-squarecorr = 19.06, P < 0.001) more com-mon in the nonfluent group.

Table 7 shows that at 4 to 12 weeks seven (12%) of thefluent aphasics had died and seven (12%) were stillmoderately or severely impaired. Of 50 survivors at 4 to 12weeks, aphasia improved in 37 (74%), and in 22 (44%) itcleared completely. Three of the seven fluent aphasics whodied had bloody CSF on admission. Six had moderate tosevere hemiparesis, and five had a visual field cut.Conversely, all those without hemiparesis and those withouta field cut survived.

Table 8 shows that at 4 to 12 weeks 38 (32%) of the non-fluent aphasics had died and 41 (34%) were still moderatelyor severely impaired. Of 82 survivors at 4 to 12 weeks,aphasia improved in 45 (55%), but in only 11 (13%) did itclear completely. Twelve of the 38 nonfluent aphasics whodied had bloody CSF. Thirty-four had moderate to severehemiparesis, and 20 had a visual field cut. (Only oneappeared to have normal visual fields; 17 were uncertain.)Conversely, the six nonfluent aphasics without hemiparesis(including one pure anomic) survived, as did 26 of the 27lacking a visual field cut.

Thus, nonfluency carried a significantly (Chi-squarecorr = 6.67, P < 0.01) worse prognosis for mortality,and within both the nonfluent and fluent groups mortalitywas associated with hemiparesis and/or visual field cut.

One nonfluent patient with a mild deficit on admissionbecame moderately impaired following a second stroke. Of

TABLE 5 Homonymous Hemianopia in Fluent and NonfiuentAphasics

FluentNonfluent

Broca'sAnomicsMixed

Total

Present

18(32%)66(55%)4(44%)3(75%)

59(55%)84(47%)

Absent

24(42%)27(23%)

5(56%)1(25%)

21(20%)51(29%)

Uncertain

15(26%)27(22%)

-_

27(25%)42(24%)

by guest on May 28, 2018

http://stroke.ahajournals.org/D

ownloaded from

170 STROKE VOL. 7, No. 2, MARCH-APRIL 1976

TABLE 6 Aphasia Severity on Admission

FluentNonfluent

Broca'sAnomicsMixed

Total

Severe

13(23%)71(59%)9(100%)0

62(58%)84(47%)

Moderate

16(28%)33(28%)

02(50%)

31(29%)49(28%)

Mild

28(49%)16(13%)

02(50%)

14(13%)44(25%)

the two mildly impaired patients who died, one had a secondstroke and the other a myocardial infarction.

Of the 82 survivors from the nonfluent group 52 (63%)were still nonfluent at 4 to 12 weeks. While speech becamefluent in one Broca's aphasic, there persisted mild dysar-ticulation and anomia.

Only 20 fluent and 33 nonfluent patients had reading andwriting checked. Many patients would not attempt eithertask. The great majority of those checked were impaired inboth skills. One fluent and two nonfluent patients had gross-ly normal reading and writing, but extensive testing to detectfine abnormalities was not done.

Discussion

The incidence of symptoms and signs has been reported ina number of large stroke populations. Details on languagefunction seldom have been given, however. For example,Marquardson's1 retrospective study of 769 acute strokepatients noted that 133 patients, or 33% of the immediatesurvivors, were aphasic. The only elaboration offered wasthat aphasia was usually a "mixed amnestic deficit," and"severe" in two-thirds. In 106 cases aphasia was combinedwith hemiparesis. Seventy-one of 88 patients followed im-proved in the hospital and in 14 the symptoms cleared com-pletely, but the average length of in-hospital follow-up wasnot stated. Hemiplegia was less likely to improve if accom-panied by aphasia, and aphasia tended to improve faster ifunaccompanied by hemiplegia.

Of two reports on the natural history of stroke inRochester, Minnesota, one,2 analyzing patients from 1945through 1954, did not state the incidence of aphasia. Theother,3 dealing with the years 1955 through 1969, noted that10% of survivors at six months were aphasic and that each ofthese "also had other disabilities." The incidence of aphasiaacutely was not defined, nor was there any breakdown as tothe type of language disturbance later present.

Omae's et al.4 study from Japan was limited to patientswith cerebral infarction. Aphasia was present in 23 of 98

TABLE 7 Status of Fluent Aphasics at 4 to 12 Weeks

Severe Moderate Mild Cleared Died

Severe onadmission(13)

Moderate onadmission(16)

Mild onadmission(28)

Total (57)

10 16

cases (24%), and was assessed by the short Minnesota Test.5

Modifying Schuell's classification, the authors consideredfour patients to have "simple" aphasia, three with "aphasiawith scattered lesions," six with "irreversible aphasia," twowith "partial auditory imperception," five with "mildaphasia with persisting dysarthria," and two with "severeaphasia unclassifiable" because of confusion. Further ex-planation of these types was not given.

David and Heyman's8 report of 100 consecutive cerebralinfarctions noted that 48 cases were in the carotid territory,32 were in the vertebrobasilar territory and 20 were "uncer-tain;" aphasia frequency was not given. Robinson's et al.7

study of the natural history of cerebral thrombosis inWorcester, Massachusetts, listed aphasia as a "minordeficit" (examples of severe deficit were hemiplegia or hemi-paresis), and did not give aphasia incidence, acutely orchronically. Gurdjian's et al.8 analysis of 600 stroke patientssimilarly treated aphasia in a cursory fashion.

There have been, conversely, many series of aphasics, butthey have not concerned stroke patients exclusively. Head9

based his elaborate aphasia classification on only 26patients, most of whom had sustained gunshot wounds andwere examined after their symptoms had stabilized.Investigators who have similarly studied largely or entirelyhead injury cases include Kleist,10 Goldstein,11 Schiller,12

Wepman,13 Russell and Espir,14 Hecaen and Angelergues,15

and Luria.16 Studies based mainly on stroke patients, but in-cluding cases of head trauma or neoplasm, include those ofWeisenberg and McBride,17 Alajouanine,24 Schuell, Jenkinsand Jiminez-Pabon,18 Marks, Taylor and Rusk,19 andBrown and Simonson.20 Sarno et al.21 evaluated 31 strokepatients, but deliberately selected only those severelyaffected. Moreover, aphasia studies, whether with strokepatients or otherwise, have largely been conducted weeks,months, or years after the acute insult. Weisenberg andMcBride's" patients, for example, were studied from twodays to ten years afterward, with an average of severalmonths. WepmanV3 cases were all assessed after six monthsbecause of the authors' belief that at that time spontaneousimprovement did not occur. Indeed, Benson and Gesch-wind27 have stressed the unreliability of classifying patientstoo early after an acute insult, since a particular type ofaphasia may evolve only after weeks or months.

TABLE 8 Status of Nonfluent Aphasics at 4 to IS Weeks

2(3%) 5(9%) 21(37%) 22(39%) 7(12%)

Severe onadmission

(71)Broca's

0)Moderate on

admission(35)

Anomic(2)

Mild onadmission

(14)Anomic

(2)Total (120)

Severe Moderate

19 14

4

7

1

19(16%) 22(18%)

Mild

5

1

15

2

10

1

Cleared

6

3

4

1

30(25%) 11(9%)

Died

27

1

9

2

1

38(32%)

by guest on May 28, 2018

http://stroke.ahajournals.org/D

ownloaded from

APHASIA IN ACUTE STROKE/Brust et at. 171

The present study deals solely with unselected, con-secutive stroke patients, each of whom was evaluated withinhours of acute insult. Twenty-one percent of all strokepatients had aphasia acutely, and while, not surprisingly, themajority were of mixed type, there were nonetheless a sub-stantial number of "pure forms": 5% Broca's, 6% anomia,14% Wernicke's, 6% conduction, and 4% "isolation" ortranscortical sensory.

There have been nearly as many classifications of aphasiaas there have been investigators, and often the same termhas been used differently by different workers. Mostclassifications have been based upon either a psychologicalor an anatomical approach. Following Broca's 1861reports28'29 suggesting the existence of a motor speech centerin the left inferior frontal convolution, the late nineteenthcentury was dominated by localizers. Wernicke30 in 1874classified aphasia into three types: sensory, from a lesion inthe posterior first temporal convolution (the putative"speech comprehension center"); motor, from a lesion inBroca's area; and central or "conduction" aphasia("leitungsaphasie," characterized by a paraphasia in spon-taneous speech and on repetition), from a disconnectionbetween the two. The notion that the elements of speech, in-cluding reading and writing, reside in discrete inter-connected centers was re-enforced by such workers asBastian,31 Lichtheim,32 and Broadbent.33

As early as 1864 Jackson" had warned, "Speaking is notsimply the utterance of words . . . speaking ispropositionizing," stressing that disturbances of languageshould be studied in terms of psychological phenomena, notareas of anatomical destruction. His views were in theminority until 1906, when Marie35 declared that Broca'saphasia was simply dysarthria plus a general intellectualdeficit especially affecting language. In 1913 Pick36 defined,psychologically, four stages in the transition from thought tospeech, and classified aphasia in terms of arrests along thishierarchical system. Head37 in 1926 defined aphasia as a dis-turbance of "symbolic formulation and expression." Herecognized four aphasia types: verbal ("defective power offorming words, whether for external or internal use"), syn-tactical ("lack of that perfect balance and rhythm necessaryto make the sounds uttered by the speaker easily com-prehensible"), nominal ("difficulty in appreciating thenominal significance of words"), and semantic ("want ofrecognition of the ultimate significance and intention ofwords and phrases apart from their direct meaning").38

Goldstein39 viewed aphasia as "dedifferentiation of brainperformance," and anomia in particular as "loss of abstractattitude." He recognized several varieties of aphasia."Three were expressive: peripheral motor, with preservedwriting; central motor, corresponding to Broca's aphasia;and "transcortical" motor, with relatively preserved repeti-tion. Three were receptive: pure word deafness; "cortical"sensory, corresponding to Wernicke's sensory aphasia; and"transcortical" sensory, with, again, relatively preservedrepetition. Goldstein also recognized "central" aphasia, cor-responding to Wernicke's conduction aphasia, and"amnestic" aphasia, or word-finding difficulty.

Brain" saw different varieties of aphasia as disturbancesof hierarchical "schemas," a concept he derived in part fromKant and from Bergson. Luria42 has based his aphasia

classification upon Pavlov's concept of cortical analyzers.Schuell and Jenkins," while dividing aphasia into fivecategories, saw them as quantitative variants of a singlebasic disturbance. Bay44 recognized only one kind ofaphasia, believing that Broca's type represented aphasia plusdysarthria and that Wernicke's sensory aphasia representedaphasia plus a general mental disorder.

A number of classifiers in this century, for exampleHenschen45 and Kleist,10 continued to emphasize anatomy.Nielsen,46 in an attempt to encompass within a classificationanatomical, physiological, and psychological features,developed categories such as "agnosia, auditory, temporal,verbal" and "aphasia, visual, semantic, external capsular,"and counted 87 types. In a recent review Meyer47 has pointedout that many workers considered holist, for example Head,have in fact associated particular anatomical lesions withtheir clinical subtypes. Nonetheless, systems such as Head'sare difficult to utilize clinically. As Weisenberg andMcBride48 stated, "Head's classification . . . seems superiorto (others) in theory and inferior in practice."

It should be stressed that basing the study of an aphasiapopulation on any classification is unlikely to produce newinsights into whether the system being utilized accuratelydetects the physiological or psychological similarities ordifferences within the population. Spreen49 has said, "thenosological system chosen determines the scope and amountof detail available for analysis. No such schema is likely toproduce more than a confirmation or a lack of confirmationfor the classification system used by the researcher."

Benson and Geschwind27 have observed that "the rate ofagreement on what is being described in the differentclassifications is actually very high." There are exceptions tothis statement. For example, there is no real counterpart inGeschwind's system to Head's semantic aphasia. More-over, by compartmentalizing patients on the basis of one ortwo shared signs, it is implied that such aphasias aremechanistically or physiologically the same. The extremevariability of speech output, qualitatively and quantitatively,in, for example, Broca's and Wernicke's aphasia, suggeststhis assumption may be fallacious. Geschwind's classifica-tion is based furthermore upon anatomical assumptionswhich are debatable (Brown60 has made a persuasive argu-ment against the concept of conduction aphasia as secon-dary to a lesion in the arcuate fasciculus, the putative cableconnecting Wernicke's and Broca's areas). Finally, whileGeschwind's emphasis on fluency versus nonfluency haslinguistic foundation," it is not always easy to call speecheither fluent or nonfluent.

A major advantage of Geschwind's system is the ease itlends to examination. Many test batteries have been devisedfor aphasia patients, some complex and time-consuming,and some testing more than language. Head's52 systematicseries of tests included naming common objects, color nam-ing, and simple reading, plus a clock-setting test, a "coin-bowl" test (requiring following directions involving numberand space), and a "hand, ear and eye test" (requiring thefollowing of directions and movement imitation). Gold-stein53 used specific language tests (spontaneous speech,series recitation, repetition, word-finding, auditory com-prehension, "responses to everyday questions and com-ments," and following directions of increasing complexity),

by guest on May 28, 2018

http://stroke.ahajournals.org/D

ownloaded from

172 STROKE VOL. 7, No. 2, MARCH-APRIL 1976

plus additional tests (stick and block designs, color andform-sorting) which he believed measured "abstractbehavior."

Weisenberg and McBride's54 tests included speaking,naming, repeating, understanding of spoken language,reading, writing, arithmetic, "language intelligence tests,"reproduction of verbal material, and an array of non-language tests. The average amount of time spent with eachpatient was 19 hours. An appendix to their book offered ab-breviated test batteries which could be performed in onlytwo to three hours. Similarly, Schuell, whose "MinnesotaTest"" is very time-consuming, published a short aphasiaexamination (requiring only 30 minutes) of four standar-dized parts10 (auditory disturbance, visual disturbance,speech and language disturbance, and visual and writing dis-turbance). Other manuals of standardized testing have in-cluded those of Eisenson,58 Wepman and Jones,67 Porch,68

Spreen and Benton,69 and more recently the "BostonDiagnostic Aphasia Examination" by Goodglass andKaplan.80

The system of examination in the present study requiredno lengthy manuals and could be done at the bedside in 15minutes with reasonable agreement from one examiner toanother. While the use of terms such as "mild" or "severe"was of course more subjective and less quantitative thannumerical scoring, it did succeed in separating patients intogross categories of severity and in assessing whether or notthey were improving.

While, as noted, the majority of our aphasics on admis-sion were of mixed type, more than one-third from the out-set had a definable "pure" form. Although the heterogeneitywithin each group was often considerable, the groupsdiffered sufficiently from each other to support Jakobson's81

notion (based upon linguistic analysis) that aphasia cannotbe viewed as "a unitory general disorder with the allegedlydifferent types of aphasia representing differences in quan-tity of disturbance rather than in quality."

There was no significant difference in age between fluentand nonfluent aphasics. It has been noted that aphasia inchildren is nearly always nonfluent,82 but age apparently haslittle influence in this regard once adulthood has beenreached.

The significantly greater percentage of men among non-fluent aphasics than among fluent aphasics and non-aphasics together is of interest. Male-female differences in avariety of higher cortical functions have been suspected for anumber of years.83"86 Recently, Kimura88 and McGlone andKertesz87 have suggested that the right hemisphere may bemore specialized for spatial processing in men than inwomen, who in turn may have more developed languageskills. Male-female differences in the development of spatialand language skills have recently been reviewed by Bufferyand Gray.88 The relation, if any, of such sex differences inverbal and non-verbal processing to differences in fluencybetween male and female aphasics is unclear. In our seriesthere were no significant differences in male-female ratioamong the "pure" aphasia types (Broca's, nonfluent andfluent anomia, Wernicke's, conduction, and "isolation").

The lower incidence of hypertension in our patients com-pared to those of the Framingham study,89 of whom only

15% with acute brain infarct had normal blood pressure,may reflect lack of premorbid information in many of ourcases.

The smaller incidence of hemiparesis in fluent than non-fluent patients parallels previous observations of others.61

The large number of fluent patients lacking a field cut andthe large number of Broca's aphasics showing one came as asurprise (although a recent pathological study70 suggests thearea of infarct causing Broca's aphasia extends considerablybeyond the so-called Broca's area).

Nonfluency carried a worse prognosis, for both mortalityand persisting severe deficit, than did fluency. Since non-fluent aphasia was mixed in 107 of 120 cases, and since abrain lesion associated with both anterior and posterior cor-tical signs is likely to be larger than a lesion associated withmore restricted findings, it is likely that this aspect, ratherthan any special feature of nonfluency, accounts for the poorprognosis. A similar explanation can be proposed for thepoor prognosis associated with either hemiparesis or fieldcut.

Seventy-four percent of the early surviving fluent aphasicsand 55% of the early surviving nonfluent aphasics improved.Such early improvement following acute aphasia is wellknown. More controversial, and not possible to determinefrom the present series, is how long aphasia continues to im-prove, a matter obviously relevant in assessing the value ofspeech therapy.71 Also not answerable because of the fre-quent omission of repetition testing during follow-up ex-aminations was how often one aphasia type evolved intoanother during improvement.

We did not detect, during this period, any examples ofsuch rare language disturbances as cortical anarthria, pureword deafness, or alexia without agraphia. Nor did we findaphasia with a lesion in the territory of an artery other thanthe middle cerebral. (Penfield and Roberts72 have showntransient aphasia to occur with lesions of the supplementarymotor area.)

A number of probably aphasic patients were not includedin the present series because gross dementia or stupor mademeaningful language testing impossible. To what extentlanguage dissolution per se disrupts non-language mentalfunctioning remains controversial. Certainly, many of oursevere aphasics showed impaired intellectual abilities inareas other than language.

Aphasia, occurring in 21% of 850 stroke patients overthree years, and initially moderate or severe in 75%,represents an enormous clinical problem at HarlemHospital. It is essential that all physicians dealing with suchpatients recognize aphasia and its varieties and try to com-municate this understanding to the patient and his family.Indeed, it is not unusual for aphasia, especially when unac-companied by other obvious neurological signs, to be calledpsychosis.73 One of our patients was considered schizo-phrenic until careful testing, especially of writing, revealedthe nature of his problem. Another patient, not included inthe present series because she was not properly examinedacutely, was sent to a mental hospital, and her aphasia wasappreciated only after she was sent back to Harlem Hospitalfor better blood pressure control. From the standpoint of apatient's ability to function in society, severe aphasia is more

by guest on May 28, 2018

http://stroke.ahajournals.org/D

ownloaded from

APHASIA IN ACUTE STROKE/Brust et al. 173

disabling than, for example, hemiplegia or blindness.Kurtzke74 has estimated the incidence of stroke in the

United States to be 207/100,000 per year. Thus, based upona 1970 U.S. population of approximately 210 million, it canbe estimated that roughly 400,000 new strokes will occur peryear, that at least 21% (or 84,000) of these patients will haveaphasia, and that at 4 to 12 weeks the language disturbancein 2.5% (or 10,000) will still be severe.

Acknowledgment

We thank Dr James Miller, Dr. Richard Rhee, and the Medical HouseStaff at Harlem Hospital Center, who assisted in patient evaluation. Ms. Vi-vian Dorset, RN, provided invaluable organization help. Ms. Marsha Holtand Ms. Ann Barnes assisted in typing the manuscript.

References

1. Marquardsen J: The natural history of acute cerebrovascular disease: Aretrospective study of 769 patients. Acta Neurol Scand 45 (Suppl 38),1969

2. Whisnant JP, Fitzgibbons JP, Kurland LT, et al: Natural history ofstroke in Rochester, Minnesota, 1945 through 1954. Stroke 2:11-22,1971

3. Matsumoto N, Whisnant JP, Kurland LT, et al: Natural history of strokein Rochester, Minnesota, 1955 through 1969: An extension of a previousstudy, 1945 through 1954. Stroke 4:20-29, 1973

4. Omae T, Katsuki S, Nishimaru K, et al: Clinical features of cerebral in-farction in the Japanese. J Chron Dis 21:585-606, 1969

5. Schuell H. A short examination for aphasia. Neurology 7:625-634, 19576. David NJ, Heyman A: Factors influencing the prognosis of cerebral

thrombosis and infarction due to atherosclerosis. J Chron Dis11:394-404, 1960

7. Robinson RW, Demirel M, LeBeau RJ: Natural history of cerebralthrombosis. Nine to 19-year follow-up J Chron Dis 21:221-230, 1968

8. Gurdjian ES, Lindner DM, Hardy WG, et al. Cerebrovascular disease:An analysis of 600 cases. Neurology 10:372-380, I960

9. Head H Aphasia and Kindred Disorders of Speech. New York,MacMillan, 1926, 2 volumes

10. Kleist K: Gehirnpathologie Varnehmlich auf grund der Kriegserfah-rungen. Leipzig, Barth, 1934

11. Goldstein K: Aftereffects of Brain Injuries in War, Their Evaluation andTreatment; the Application of Psychologic Methods in the Clinic. NewYork, Grune & Stratton, 1942

12. Schiller J: Aphasia studied in patients with missile wounds. J NeurolNeurosurg Psychiat 10:183-197, 1947

13. Wepman JM: Recovery From Aphasia. New York, Ronald Press Co,1951

14. Russell WR, Espir MLE: Traumatic Aphasia; A Study of Aphasia inWar Wounds of the Brain. London, Oxford University Press, .1961

15. Hecaen H, Angelergues R: Localization of symptoms in aphasia. In deReuck AVS, O'Connor M (eds): Disorders of Language. Boston, Little,Brown Co, pp 223-256, 1964

16. Luria AR: Traumatic Aphasia; Its Syndromes, Psychology and Treat-ment. Mouton, The Hague, 1970

17. Weisenberg T, McBride KE: Aphasia. A Clinical and PsychologicalStudy. New York, The Commonwealth Fund, 1935

18. Schuell H, Jenkins JJ, Jimenez-Pabon E: Aphasia in Adults. New York,Hoeber, 1964

19. Marks MM, Taylor ML, Rusk HA: Rehabilitation of the aphasicpatient: A survey of 3 years' experience in rehabilitation setting.Neurology 7:837-843, 1947

20. Brown JR, Simonson J: A clinical study of 100 aphasic patients.Neurology 7:777-783, 1957

21. Sarno MT, Silverman M, Sands E: Speech therapy and languagerecovery in severe aphasia. J Speech Hearing Res 13:607-623, 1970

22. Shafer SQ, Bruun B, Richter RW: Brain infarction risk factors in blackNew York City stroke patients. J Chron Dis 27:127-133, 1974

23. Geschwind N: Aphasia. New Eng J Med 284:654-656, 197124. Alajouanine T: Verbal realization in aphasia. Brain 79:1-28, 195625. Critchley M (ed): The nature and content of aphasic utterance. In

Aphasiology and Other Aspects of Language. London, Edward ArnoldLtd, pp 227-234, 1970

26. Goldstein K: Die transkortikalen Aphasien. Jena, Gustav Fischer, 191727. Benson DF, Geschwind N: The aphasias and related disturbances. In

Baker AB, Baker LH (eds): Clinical Neurology. Volume 1, part 8, NewYork, Harper & Row, 1971

28. Broca PP: Perte de la parole. Ramollissement chronique et destructionpartielle du lobe anterieur guache du cerveau. Bull Soc Anthrop (Paris)2:235-238, 1861

29. Broca PP: Remarques sur la siege de la faculte du language articule,suives d'une observation d'aphemie. Bull Soc Anat (Paris) 6:330-357,1861

30. Wernicke C: Der aphasische Symptomcomplex: Eine psychologischeStudie auf anatomischer Basis. Breslau, Cohn and Weigert, 1874

31 Bastian HC On the various forms of loss of speech in cerebral disease.Brit Foreign Med Chir Rev (London) 43:209-236, 470-492, 1869

32 Lichtheim L On aphasia. Brain 7:433-484, 188533. Broadbent WH: On the cerebral mechanism of speech and thought.

Trans Roy Med Chir Soc (London) 55:145-194, 187234. Jackson JH- Selected Writings. Volume 2. London, Hodder and

Stoughton, p 121, 193235. Marie P: Revision de la question de I'aphasie. Sem Med 26:241-247,

493-500, 565-571, 190636. Pick A: Die agrammatischen Sprachstorungen. Berlin, Springer, 191337. Head H. 1926, op cit, pp 210-21238. Ibid, pp 221-26839. Goldstein K: Language and Language Disturbances. New York, Grune

and Stratton, p 56, 194840. Ibid, pp 190-32441. Brain WR Speech Disorders. Aphasia, Apraxia and Agnosia. London,

Butterworths, pp 67-81, 196542 Luria AR: Factors and forms of aphasia In de Reuck AVS, O'Connor

M (eds): Disorders of Language. Boston, Little, Brown Co, pp 143-167,1964

43. Schuell H, Jenkins JJ: The nature of language deficit in aphasia. PsycholRev 66:45-47, 1959

44. Bay E: Principles of classification and their influence on our concepts ofaphasia. In de Reuck AVS, O'Connor M (eds). Disorders of Language.Boston, Little, Brown Co, pp 122-142, 1964

45 Henschen SE: Klimsche und pathologische Beitrage zur Pathologie desGehirns, Vols V, VI and VIII. Stockholm, Nardiske Bokhandeln,1920-1922

46. Nielsen JM' Agnosia, Apraxia, Aphasia. Their Value in CerebralLocalization, (appendix) New York, Hoeber, 1946

47. Meyer A: The frontal lobe syndrome, the aphasias, and related con-ditions — a contribution to the history of cortical localization. Brain97:565-601, 1974

48. Weisenberg T, McBride KE: 1935, op cit, p 5149. Spreen O: Language disturbances in cerebrovascular disease. In Benton

AL (ed): Behavioral Change in Cerebrovascular Disease. New York,Harper & Row, pp 40-45, 1970

50. Brown JW: Aphasia, Apraxia, Agnosia. Clinical and TheoreticalAspects. Springfield, Illinois, Charles C Thomas, pp 77-101, 1972

51. Howes D: Application of the word-frequency concept to aphasia. In deReuck AVS, O'Connor M (eds): Disorders of Language. Boston, Little,Brown Co, pp 47-78, 1964

52. Head H. 1926, op cit, pp 145-16553. Goldstein K 1948, op cit, pp 153-18854. Weisenberg T, McBride KE: 1935, op cit, pp 135-13655. Schuell HM: Minnesota Test for Differential Diagnosis of Aphasia.

Research Edition. Minneapolis, University of Minnesota Press, 195556. Eisenson J: Examining for Aphasia. New York, Psychological Corp,

195457. Wepman JM, Jones LV: The language modalities test for aphasia.

Chicago, The Industrial Relations Center, University of Chicago, 196158. Porch BE: Porch Index of Communicative Ability. Palo Alto, Consulting

Psychologist Press, 196759. Spreen O, Benton AL: Neurosensory Center Comprehensive Examina-

tion for Aphasia. Canada, University of Victoria, 196860. Goodglass H, Kaplan E: The Assessment of Aphasia and Related

Disorders. Philadelphia, Lea and Febiger, 197261. Jakobson R: Toward a linguistic typology of aphasia impairments. In de

Reuck AVS, O'Connor M (eds): Disorders of Language. Boston, Little,Brown Co, pp 21-46, 1964

62. Alajouanine T, Lhermitte F: Acquired aphasia in children. Brain88:653-662, 1965

63. Hobson J: Sex differences in primary mental abilities. J Ed Psych41:126-132, 1974

64. Meyer W, Bendig A: A longitudinal study of the Primary MentalAbilities Test. J Ed Psych 52:50-60, 1961

65. Landsdell H: A sex difference in effect of temporal-lobe neurosurgery ondesign preference. Nature 194:852-854, 1962

66. Kimura D: Spatial localization in left and right visual fields. Canad JPsychol 23:445-458, 1969

67. McGlone J, Kertesz A: Sex differences in cerebral processing of visuo-spatial tasks. Cortex 9:313-320, 1973

68. Buffery AWH, Gray JA: Sex differences in the development of spatial

by guest on May 28, 2018

http://stroke.ahajournals.org/D

ownloaded from

174 STROKE VOL. 7, No. 2, MARCH-APRIL 1976

and linguistic skills. In Ounsted C, Taylor DC (eds): Gender Differences:Their Ontogeny and Significance. New York, Longman, pp 123-157,1972

69. Kannel WB, Wolf PA, Verter J, et al: Epidemiologic assessment of therole of blood pressure in stroke. The Framingham study. JAMA214:301-310, 1970

70. Mohr JP, Funkenstein HH, Finkelstein S, et al: Broca's area infarctionversus Broca's aphasia, (abstr) Neurology 25:349, 1975

71. Darley FL: Treatment of acquired aphasia. In Friedlander WJ (ed): Ad-vances in Neurology, Vol 7. New York, Raven Press, pp 111-145, 1975

72. Penfield W, Roberts L: Speech and Brain Mechanisms. Princeton,Princeton University Press, p 201, 1959

73. Critchley M: The neurology of psychotic speech. Br J Psychiat110:353-364, 1964

74. Kurtzke JF: Epidemiology of Cerebrovascular Disease. Berlin, Springer-Verlag, p 148, 1969

An In Vitro Study of Prolonged Vasospasmof a Rabbit Cerebral Artery

SUE PIPER DUCKLES, PH.D., ROSEMARY D. BEVAN, M.D., AND JOHN A. BEVAN, M.D.

SUMMARY Longitudinal stretch of the rabbit basilar arteryproduces local injury followed by prolonged circular constriction.After stretching and rapid release in vitro localized constrictionspromptly occurred. This could be prevented by prior treatment withcyanide or calcium-free solution. Once produced, constrictions per-sisted for more than 72 hours. Previously induced constriction wasnot reversed by treatment for two hours with cyanide or by removingcalcium. Histological observation indicated that constricted areaswere associated with a discrete circumferential rupture of the internalelastic lamina and disruption and thinning of the underlying media.

Specific catecholamine fluorescence at the adventitio-medial junctionwas unchanged in constricted areas. The relationship between smoothmuscle cell length and resting tension of artery segments with andwithout constrictions was compared. Segments with constrictions hada shorter muscle length for any given resting tension, which confirmsthat constriction was not due to passive collapse of the vessel wall.These findings suggest that injury of cerebrovascular smooth musclemay result in essentially irreversible vasoconstriction. Such a mecha-nism could contribute to the pathogenesis of prolonged cerebral vaso-spasm after SAH or traumatic injury to the cerebrum.

MOST WORKERS agree that release of vasoactive sub-stances from subarachnoid blood is a major cause of diffusecerebral vasospasm after rupture of an intracranial aneu-rysm.'"4 Yet there is experimental evidence that spasm maybe more prolonged and severe when puncture of a cerebralvessel is combined with injection of blood into the sub-arachnoid space than when blood is applied without in-jury ."•" Studies of mechanical stimulation of cerebral vesselshave found the resulting constriction to be short lived,lasting less than 30 minutes.7"9 It is this failure todemonstrate prolonged spasm after mechanical stimulationof or injury to cerebral vessels that has led to the conclusionthat these factors cannot contribute to the etiology ofcerebral vasospasm.2

The present paper demonstrates that longitudinal stretchof the rabbit basilar artery can produce local injury followedby prolonged discrete circular constriction. The accom-panying histological changes are described, and the role ofsmooth muscle in the production of constriction is con-firmed. These findings demonstrate that under certain con-ditions, such as injury, cerebral arterial smooth muscle mayundergo an essentially irreversible contraction. Such amechanism could contribute to the pathogenesis and patho-

From the Departments of Pharmacology and Pathology, University ofCalifornia School of Medicine, Center for Health Sciences, Los Angeles,California 90024.

This investigation was supported by the following grants: US Public HealthService Grant #HL 15805 to Dr. John Bevan and American HeartAssociation-Greater Los Angeles Affiliate Grant # 408 IG. Dr. Duckies issupported by an NIH Fellowship f IF 22 NS 03104-01.

physiology of prolonged cerebral vasospasm after sub-arachnoid hemorrhage or traumatic injury.

A preliminary report of a portion of this work has beenpresented at the Second International Symposium onVascular Neuroeffector Mechanisms, Odense, Denmark.10

Methods

New Zealand white rabbits (weight 2 to 3 kg) werestunned by a blow on the nose and bled from the neck. Theentire brain with attached arachnoid membrane and bloodvessels was removed and placed in Krebs bicarbonate solu-tion at room temperature containing (mM): Na+, 144.2; K+,4.9; Ca++, 1.3; Mg++, 1.2; Cl, 126.7; HCOi, 25.0; SO;,1.19; ethylene diamine tetraacetic acid, 0.027; and glucose,11, and bubbled with 95% O2 and 5% CO2. The basilar arterywas observed and further dissection was carried out using aWild stereomicroscope. Photographs were taken with a 35-mm camera attachment. In some experiments Na CN (200mg per liter) was added to the Krebs solution and glucosewas omitted. In other experiments calcium was omittedfrom the Krebs solution and 2 mM EGTA (ethyleneglycol-bis [/3-amino ethyl ether] N,N-tetraacetic acid) was added.

HISTOLOGY

After dissection and diagrammatic recording of thepresence and site of areas of constriction, the basilar arterywas cut open longitudinally and laid flat on a small card.This was then fixed in buffered formaldehyde, dehydrated

by guest on May 28, 2018

http://stroke.ahajournals.org/D

ownloaded from

J C Brust, S Q Shafer, R W Richter and B BruunAphasia in acute stroke.

Print ISSN: 0039-2499. Online ISSN: 1524-4628 Copyright © 1976 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Stroke doi: 10.1161/01.STR.7.2.167

1976;7:167-174Stroke. 

http://stroke.ahajournals.org/content/7/2/167World Wide Web at:

The online version of this article, along with updated information and services, is located on the

  http://stroke.ahajournals.org//subscriptions/

is online at: Stroke Information about subscribing to Subscriptions: 

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

  document. Permissions and Rights Question and Answer available in the

Permissions in the middle column of the Web page under Services. Further information about this process isOnce the online version of the published article for which permission is being requested is located, click Request

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.Stroke Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:

by guest on May 28, 2018

http://stroke.ahajournals.org/D

ownloaded from